铁路轨道异物完整性检测与跟踪算法研究
|
摘要
针对野外复杂环境下轨道异物检测不完整问题,提出基于小波变换的像素过滤思想改进GMM,构建背景模型;为解决异物目标实施机动(转弯、加速或突然出现)时跟踪实时性差和准确率低的问题,分析Kalman滤波线性化误差,搭建BP神经网络修正IMM的跟踪模型,实现轨道异物跟踪预测,并推导出非线性Kalman滤波关系.实验表明,改进GMM在正常天气下平均前景误检率降低了24.94个百分点,针对复杂恶劣天气平均前景误检率降低了33.76个百分点;建立BP神经网络-IMM-Kalman滤波模型不仅可以快速准确地对场景中的机动目标进行跟踪,而且比Kalman滤波和IMM更加平稳,误差更小.
Aiming at the problem of incomplete detection of orbital objects in complex field environment, this paper proposes a pixel filtering idea based on wavelet transform to improve GMM and construct a background model. The maneuvering(turning, accelerating or sudden appearance) is implemented to solve the foreign object target. When tracking the problem of low real-time performance and low accuracy, analyzing the linearization error of Kalman filter, BP neural network is used to modify the tracking model of IMM to realize the tracking and prediction, and the nonlinear Kalman filtering relationship is derived. The experimental shows that the average foreground false detection rate of the improved GMM is reduced by 24.94 percentage points under normal weather conditions, and the average prospect false detection rate for complex bad weather is reduced by 33.76 percentage points. The BP neural network-IMM-Kalman filtering model is not only established. It can track the maneuvering target in the scene quickly and accurately, and it is more stable and less error than the Kalman filtering algorithm and IMM.
Aiming at the problem of incomplete detection of orbital objects in complex field environment, this paper proposes a pixel filtering idea based on wavelet transform to improve GMM and construct a background model. The maneuvering(turning, accelerating or sudden appearance) is implemented to solve the foreign object target. When tracking the problem of low real-time performance and low accuracy, analyzing the linearization error of Kalman filter, BP neural network is used to modify the tracking model of IMM to realize the tracking and prediction, and the nonlinear Kalman filtering relationship is derived. The experimental shows that the average foreground false detection rate of the improved GMM is reduced by 24.94 percentage points under normal weather conditions, and the average prospect false detection rate for complex bad weather is reduced by 33.76 percentage points. The BP neural network-IMM-Kalman filtering model is not only established. It can track the maneuvering target in the scene quickly and accurately, and it is more stable and less error than the Kalman filtering algorithm and IMM.
引文
[1]衷路生,李兵,龚锦红,等.高速列车非线性模型的极大似然辨识[J].自动化学报,2014,40(12):2950-2958.[ZHONG L S,LI B,GONG J H,et al.Maximum likelihood identification of nonlinear model for highspeed train[J].Acta Automatica Sinica,2014,40(12):2950-2958.]
[2]王迪,陈光武,杨厅.一种快速高精度GPS组合定位方法研究[J].铁道学报,2017,39(2):67-73.[WANG D,CHEN G W,YANG T.Study on a fast and precise GPSintegrated positioning method[J].Journal of the China Railway Society,2017,39(2):67-73.]
[3]THIERRY BOUWMANS,FIDA EL BAF,BERTRANDVACHON.Background modeling using mixture of gaussians for foreground detection a survey[J].Recent Patents on Computer Science,2008,1(3):219-237.
[4]SHEIKH Y,SHAH M.Bayesian modeling of dynamic scenes for object detection[J].IEEE Transactions on Patttern Analysis and Machine Intelligence,2005(27):1778-1792.
[5]林庆,徐柱,王士同,等.HSV自适应混合高斯模型的运动目标检测[J].计算机科学,2010,37(10):254-256,290.[LIN Q,XU Z,WANG S T,et al.Moving objects detection of adaptive Gaussian mixture models on HSV[J].Computer Science,2010,37(10):254-256,290.]
[6]刘静,王玲.混合高斯模型背景法的一种改进算法[J].计算机工程与应用,2010,46(13):168-170.[LIU J,WANG L.An improved algorithm of Gaussian mixture model background method[J].Computer Engineering and Applications,2010,46(13):168-170.]
[7]华媛蕾,刘万军.改进混合高斯模型的运动目标检测算法[J].计算机应用,2014,34(2):580-584.[HUA YL,LIU W J.Moving object detection algorithm of improved Gaussian mixture model[J].Journal of Computer Applications,2014,34(2):580-584.]
[8]尹宏鹏,陈波,柴毅,等.基于视觉的目标检测与跟踪综述[J].自动化学报,2016,42(10):1466-1489.[YINH P,CHEN B,CHAI Y,et al.Vision-based object detection and tracking:A review[J].Acta Automatica Sinica,2016,42(10):1466-1489.]
[9]CHANG DAHCHUNG,FANG MENGWEI.Bearingonly maneuvering mobile tracking with nonlinear filtering algorithms in wireless sensor networks[J].IEEETrans on Systems Journal,2014,8(1):160-170.
[10]张云洲,付文艳,项姝,等.室内环境下基于IMM-EKF算法的移动目标定位[J].计算机研究与发展,2014,51(11):2408-2415.[ZHANG Y Z,FU W Y,XIANG S,et al.IMM-EKF algorithm-based indoor moving target localization[J].Journal of Computer Research and Development,2014,51(11):2408-2415.]
[11]宫淑丽,王帮峰,吴红兰,等.基于IMM算法的机场场面运动目标跟踪[J].系统工程与电子技术,2011,33(10):2322-2326.[GONG S L,WANG B F,WU H L,et al.Tracking of moving targets on airport surface based on IMM algorithm[J].Systems Engineering and Electronics,2011,33(10):2322-2326.]
[12]宋青平,刘荣科.自适应融合IMM的自主无线电载波跟踪算法[J].宇航学报,2015,36(5):605-612.[SONGQ P,LIU R K.An adaptive fusion IMM algorithm for autonomous radio carrier tracking[J].Journal of Astronautics,2015,36(5):605-612.]
[13]崔留争,高思远,贾宏光,等.神经网络辅助卡尔曼滤波在组合导航中的应用[J].光学精密工程,2014,22(5):1304-1311.[CUI L Z,GAO S Y,JIA H G,et al.Application of neural network aided Kalman filtering to SINS/GPS[J].Optics and Precision Engineering,2014,22(5):1304-1311.]
[14]谭兴龙,王坚,赵长胜.神经网络辅助的GPS/INS组合导航自适应UKF算法[J].测绘学报,2015,44(4):384-391.[TAN X L,WANG J,ZHAO C S.Neural network aided adaptive UKF algorithm for UPS/INS integration navigation[J].Acta Geodaetica et Cartographica Sinica,2015,44(4):384-391.
[15]CHANG S G,YU B,VETTERLI M.Adaptive wavelet thresholding for image de-noising and compression[J].IEEE Transactions on Image Processing,IEEETransaction on,2000,9(9):1522-1546.
[16]ALI N H,HASSAN G M.Kalman filter tracking[J].International Journal of Computer Applications,2014,89(9):15-18.
[2]王迪,陈光武,杨厅.一种快速高精度GPS组合定位方法研究[J].铁道学报,2017,39(2):67-73.[WANG D,CHEN G W,YANG T.Study on a fast and precise GPSintegrated positioning method[J].Journal of the China Railway Society,2017,39(2):67-73.]
[3]THIERRY BOUWMANS,FIDA EL BAF,BERTRANDVACHON.Background modeling using mixture of gaussians for foreground detection a survey[J].Recent Patents on Computer Science,2008,1(3):219-237.
[4]SHEIKH Y,SHAH M.Bayesian modeling of dynamic scenes for object detection[J].IEEE Transactions on Patttern Analysis and Machine Intelligence,2005(27):1778-1792.
[5]林庆,徐柱,王士同,等.HSV自适应混合高斯模型的运动目标检测[J].计算机科学,2010,37(10):254-256,290.[LIN Q,XU Z,WANG S T,et al.Moving objects detection of adaptive Gaussian mixture models on HSV[J].Computer Science,2010,37(10):254-256,290.]
[6]刘静,王玲.混合高斯模型背景法的一种改进算法[J].计算机工程与应用,2010,46(13):168-170.[LIU J,WANG L.An improved algorithm of Gaussian mixture model background method[J].Computer Engineering and Applications,2010,46(13):168-170.]
[7]华媛蕾,刘万军.改进混合高斯模型的运动目标检测算法[J].计算机应用,2014,34(2):580-584.[HUA YL,LIU W J.Moving object detection algorithm of improved Gaussian mixture model[J].Journal of Computer Applications,2014,34(2):580-584.]
[8]尹宏鹏,陈波,柴毅,等.基于视觉的目标检测与跟踪综述[J].自动化学报,2016,42(10):1466-1489.[YINH P,CHEN B,CHAI Y,et al.Vision-based object detection and tracking:A review[J].Acta Automatica Sinica,2016,42(10):1466-1489.]
[9]CHANG DAHCHUNG,FANG MENGWEI.Bearingonly maneuvering mobile tracking with nonlinear filtering algorithms in wireless sensor networks[J].IEEETrans on Systems Journal,2014,8(1):160-170.
[10]张云洲,付文艳,项姝,等.室内环境下基于IMM-EKF算法的移动目标定位[J].计算机研究与发展,2014,51(11):2408-2415.[ZHANG Y Z,FU W Y,XIANG S,et al.IMM-EKF algorithm-based indoor moving target localization[J].Journal of Computer Research and Development,2014,51(11):2408-2415.]
[11]宫淑丽,王帮峰,吴红兰,等.基于IMM算法的机场场面运动目标跟踪[J].系统工程与电子技术,2011,33(10):2322-2326.[GONG S L,WANG B F,WU H L,et al.Tracking of moving targets on airport surface based on IMM algorithm[J].Systems Engineering and Electronics,2011,33(10):2322-2326.]
[12]宋青平,刘荣科.自适应融合IMM的自主无线电载波跟踪算法[J].宇航学报,2015,36(5):605-612.[SONGQ P,LIU R K.An adaptive fusion IMM algorithm for autonomous radio carrier tracking[J].Journal of Astronautics,2015,36(5):605-612.]
[13]崔留争,高思远,贾宏光,等.神经网络辅助卡尔曼滤波在组合导航中的应用[J].光学精密工程,2014,22(5):1304-1311.[CUI L Z,GAO S Y,JIA H G,et al.Application of neural network aided Kalman filtering to SINS/GPS[J].Optics and Precision Engineering,2014,22(5):1304-1311.]
[14]谭兴龙,王坚,赵长胜.神经网络辅助的GPS/INS组合导航自适应UKF算法[J].测绘学报,2015,44(4):384-391.[TAN X L,WANG J,ZHAO C S.Neural network aided adaptive UKF algorithm for UPS/INS integration navigation[J].Acta Geodaetica et Cartographica Sinica,2015,44(4):384-391.
[15]CHANG S G,YU B,VETTERLI M.Adaptive wavelet thresholding for image de-noising and compression[J].IEEE Transactions on Image Processing,IEEETransaction on,2000,9(9):1522-1546.
[16]ALI N H,HASSAN G M.Kalman filter tracking[J].International Journal of Computer Applications,2014,89(9):15-18.